Analysis Cryptocurrency Prediction Price Using Recurrent Neural Network (RNN) Gate Recurrent Unit (GRU) Long Short-Term Memory (LSTM)

Abstract

Cryptocurrencies, which are digital assets intended to function as a medium of electronic exchange, have garnered significant global attention, with Bitcoin standing out as the most recognized example. A variety of neural network models, including Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU), have been employed in the quest to predict Bitcoin’s price movements. Numerous experiments have been conducted using different training configurations, specifically with epochs set at $100,200,300$, and 500, along with batch sizes of 32,64, and 128. These tests have consistently demonstrated that the GRU model surpasses both the RNN and LSTM models in terms of prediction accuracy and overall consistency. The results indicate that GRU presents a more reliable framework for forecasting trends in Bitcoin prices, offering greater stability and precision. Consequently, the GRU model is emerging as a preferred choice for those aiming to refine predictive analytics within the cryptocurrency market.